Operators are moving toward higher-efficiency filters for their
Operators are moving toward higher-efficiency filters
Credit: H. Johnstone

One of the most important components of a gas turbine power plant in terms of ensuring high performance is the air inlet filtration system. David Appleyard explores the latest filtration advances and how the technology will evolve to meet the changing market demands.

Small they may be, but airborne particulates nonetheless present a significant commercial threat to the owners and operators of gas turbines. Effective air inlet filtration is the obvious solution, but this inevitably presents a number of technical challenges.

One issue is the diverse nature of airborne material. Depending on their size, frequency and composition, such particulates have three broad spheres of influence, namely erosion, corrosion and fouling. Larger particles, above 5 microns, are considered to be mainly responsible for erosion while for corrosion and fouling, smaller particles of one micron or less are more influential.

In any event, as Jim Benson, senior product engineer for Camfil Farr Power Systems in North America, explains, in addition to the dust load, there are also industrial contaminants to contend with, such as hydrocarbons, cement, shot-blast sand, dust from construction work and so on. The climate also brings unwanted conditions including salt-laden coastal storms or frequent ground fog.

Each of these factors – which chiefly affect the compressor stages but can also degrade components deep within the turbine – can have a deleterious effect on the performance of the machine resulting in higher operating costs and expensive shutdowns.

On-line or off-line washing can also present challenges, notably a shutdown potentially costing millions in lost production, plus costs associated with labour, water and chemicals in the wash process. Furthermore, on-line flushing may only clean blades on the first through to the third stage, but not much further. Therefore off-line washing is much more effective and is generally required to maintain compressor performance.

Given that many gas turbines operate in baseload mode, and are thus running upwards of 6000-8000 hours a year, even a small drop in performance quickly adds up to a significant volume of lost energy and a measurable price per MWh.

Indeed, Gene Schockemoehl, president of Braden Manufacturing, explains that air filtration plays a vital role in the efficiency of the gas turbine output and the long-term care and maintenance of the turbine itself. Extending turbine life is something which operators are becoming more interested in, too, he adds.

Of course, in order to provide the best possible protection and optimise engine performance, the filtration solution must be adapted to the specific site conditions and the filter media should be chosen to give the lowest life-cycle cost.

Types of filtration system

Filter house systems come in two broad configurations: self-cleaning pulse type systems, which use compressed air to blow contaminants out of the filter media; and static, barrier types that use consumable media.

Pulse systems use round filters and were designed for relatively dry regions with a high dust loading such as the Middle East, while static systems use rectangular filters and were designed for lower concentration areas including coastal and offshore applications, were emphasis is on higher efficiencies and not necessarily the ability to self-clean.

In Europe, filters are rated according to their performance, so a basic filter may be rated at F7 through to F9. Sequentially following on from F9 higher efficiency media start at E10 through E12, and then move into H or high grade for High Efficiency Particulate Air (HEPA) filters.

Bryan Xu, product manager at Nordic Air Filtration/TDC Filter, says: “Using F6 filters in an air inlet with 480 sets of cylindrical & conical filters, where approximately 300 kg/dust enters annually, you may still get 8 kg of dust entering your turbine per year, but by using F9 filters, you will reduce the amount to 1 kg per year. If you choose ultra-high efficiency HEPA12 filters, the amount of dust goes down to 0.01 kg per year.”

The United States uses a different system – from the American Society of Heating, Refrigeration and Air Conditioning Engineers (Ashrae) – though the two broadly coincide in respect of their overall testing and objectives.

The amount of dust entering the turbine by filter class
The amount of dust entering the turbine by filter class
Credit: Midwesco/TDC Filter

From the basic cellulosic and glass fibre materials there is cellulose blended with about 20 per cent synthetic through to 100 per cent synthetic material giving a longer life and durability, and resistance to moisture.

Then additional treatments can be applied to base filter media such as nanotechnology treatment or a PTFE bath treatment to improve filter performance. More recently, membrane filters have been introduced. Their smaller pore sizes work very well in aggressive environments with high moisture concentration.

However, there are inevitably trade-offs. More efficient media are inevitably going to cost more money and may magnify the influence of negative factors, such as increasing the pressure differential across the filter.

Thus choosing the correct media is really a question of economics and considering the life-cycle costs, as well as the procurement price for the filters, their efficiency performance and their service life. For a gas turbine, a typical lifespan for a filter house or filter elements can range from 18 months to four years, depending on the environment.

Benson explains: “If you use what I’ll call an entry level or commodity filter a typical operator would have to go off line once or twice a year in order to clean their turbine blades and they would lose 3-4 per cent of their power output. If you install a higher level of filter efficiency they no longer have to go off line as frequently and have greater power output.

“It is a balancing act when an operator considers their filter selection. Factors that need to be considered, in addition to the filter’s cost, include the filter’s efficiency, pressure drop, and service life. And how those factors will impact the performance of the turbine and an operator’s maintenance costs; for example reduced compressor washes. However, It’s not of great value for an operator if you get very high-efficiency filters and they have to replace them every three months, so you need to find the optimisation between life efficiency and economics.”

Trends in filtration

There has been increasing emphasis towards higher filtration efficiency levels recently, given their advantages for operators in that they prevent the ingress of sub-micron particulates. To give a perspective on the range of particles being captured by the higher-efficiency filters: E10 are initially capturing about 85 per cent of 0.3 micron particles and above, a similar size range to tobacco smoke, says Benson.

CamGT is a compact high-efficiency, high-capacity filter
CamGT is a compact high-efficiency, high-capacity filter
Credit: Camfil Farr Power Systems

As previously mentioned; it is these smaller size dust particles and salt particles which are responsible for much of the corrosion and fouling. Xu further explains that it is sub-micron particles which accounts for the largest fraction of particulates in terms of number in typical atmosphere air.

Benson highlights this change saying: “If you were to look at filters years ago they would typically be an F7 range. Today the F9 is essentially a bare minimum with interest focusing on higher efficiency levels in the E grades from the E10s to the E12s.

“Operators are realising that you can improve machine performance by selecting the right filtration, as opposed to say 20 years ago when they only wanted to prevent catastrophic damage. That’s why there’s a much greater level interest on filtration performance because it can impact their performance, their economics.”

Mike Garnett, global market manager for Turbomachinery at Freudenberg Filter,s echoes this, saying: “About five years ago we began to see a market trend towards HEPA filtration, which is E10, 11, 12, and even H13. Before [then] it was F8, F9, which is pretty much a standard filtration sequence.”

He adds this trend has been growing ever since “we converted several hundred users of our filters and not a single one has gone back to the lower grades. They would sacrifice, even if you have to have a higher differential pressure, to have a cleaner engine/compressor The benefits in kind are about 5 to 1”.

“Even though you can calculate the higher presser differential loss, the savings you make by having a clean compressor outweigh having that slightly higher differential pressure.”

Given the benefits of higher-efficiency filtration there are a number of operational design considerations necessary to accommodate it most effectively. The cross-sectional area of the intake air path is very important, for example, with Schockemoehl noting: “In many applications with high performance filters and inlet cooling, the intake systems are much larger with increased cross-sectional area to achieve low pressure drop and increased gas turbine performance.”

Garnett expands on this, saying: “People are becoming much more aware that if you want to use high-efficiency filters you may need to reduce the airflow to lower the differential pressure. On top of that companies are also trying to introduce filters into the market with higher surface areas and greater depth to increase the surface area up so we can reduce the pressure drop. Filters originally maybe had 18 or 20 m2 of media and now we’ve introduced filters with say 30 m2.”

There are other influences at work here too though. For example, some systems have as many as three filtration stages, but more efficient filters may reduce the need for as many stages, balancing out the pressure differential. Furthermore, new synthetic materials are reducing this comparative pressure drop, with as low or even lower initial pressure drops than conventional filters. The result is low operational pressure drops during periods with high humidity, fog, rain and hurricanes to allow constant high power output of the power plants, reported in new materials by Xu.

Improving performance

Although traditional turbine OEMs are conservative and typically only specify standard filtration, which works well at eliminating larger erosive particles, there is evidence of a move towards higher-efficiency filters.

“It’s continually growing, we’ve seen OEMs like the GEs, Siemens and Alstoms now starting to change their specifications from two or three years ago, they are now introducing the options for higher efficiency filtration, cleaner compressors,” says Garnet.

Filter manufacturers are also working with turbine companies to look at the advantages and differentiations between commodity systems and their more sophisticated cousins. But there is also evidence that owners and operators are perceiving the benefits of advanced filtration more clearly too.

“I think for new systems there are occasions when the end customer is more assertive and specifying the system they need for their environment and so specifying above entry level or bare minimum requirements,” says Benson.

Nordic Air Filtration/TDC Filter is working with a major utility, which lost about 4-5 MW on a baseload machine using off-line water washing and F7/F8 filters on its Frame 7 turbine.

“By upgrading to F9 (MERV 16) high-efficiency filters, these losses were halved between off-line washes. In other words, they are able to actually recover about 2-2.5 MW by using high-efficiency filters. Now that they have seen the improved performance, they have decided to move to HEPA12/E12 filters so that they can fully recover the 4-5 MW between the off-line water washes, says Xu.

“Now the industry understands the benefits of higher-efficiency filters. Not only the plant operators, but now also the managers and the financial planning team are realising why they need to invest in these filters.”

There is also a push for more accurate analysis of environmental conditions to inform the filter system design.

For example, Benson says: “We have a [number of] mobile air filter test labs. We can test up to four separate filter systems or variants at once, so we’d bring the lab to the actual application and draw air flow through it on a single outlet basis.

“There’s a lot to be learned by testing the filters in the actual environment. It’s not going to have standard test particulate, it’s going to have hydrocarbons and salts and sands and insects etc. So having the test rig on site for four months will start to give us some trend analysis for the environment, which will give you an indication of service life.”

Looking at new advances, Schockemoehl explains: “We’re developing new technologies in terms of higher efficiency and higher performance with the thought in mind that it provides longer-term protection for the gas turbine. More and more machines are being designed with micro-particle filtration applications.”

On the issue of technological advances, Garnett says: “The next thing I think is to try and keep the high-efficiency filtration and strive to increase the dust holding and decrease the differential pressure, while at the same time keep it competitive. There are various different media and materials out on the market that offer certain aspects and benefits; it’s more evolution in this market than revolution.”

Braden is also working on ways to improve constructability of filtration systems with some new designs looking at how to make the installation easy in the field for owner/operators and installers.

This, says Schockemoehl, is increasingly becoming part of the evaluation process when considering a retrofit or new installation given that air filtration systems are relatively complex with mechanical, electrical and structural disciplines involved. Thus, part of the next generation of installations is to focus on that complexity, with more plug and play connections for example so that installation in the field is much easier.

Garnett notes that even with these considerations, the switch to high efficiency is yielding benefits: “They’ve all moved that way. The customers are also starting to mature and there’s been a lot of discussions over the last two or three years about this. I think the more mature customers have changed over one or two units and now when they order new units from the manufacturers they will probably ask for the HEPA option to look at and the cost calculations right at the beginning.”

Many air filtration system manufacturers also look at other aspects of inlet air treatment such as heating or more commonly cooling.

Braden, for instance, produces both evaporative cooling systems and chilling systems which use coils in the airflow to control temperature. Schockemoehl says: “Each system is normally uniquely designed, taking into account the environmental loading and site conditions: it can be quite challenging to ensure that all these parameters are met.” He highlights the “significant number of opportunities for enhancement and upgrading systems in the Middle East that are in place right now”.

Freudenberg Filter also produces a bespoke system, called VAM, which is a vapour absorption machine, which uses any waste heat.

Xu notes a trend in which some of their customers are converting from evaporative systems to fogging systems to achieve true HEPA efficiency in the inlet air cooling and filtration systems.

The future market

With both manufacturers and owner/operators becoming more aware of advances in high-efficiency filtration and the potential economic benefits that may bring, it seems that the trend towards HEPA systems will continue.

As Benson says: “I’m not sure if the improvements in filtration drove the recognition of better economics for the operator or the need for better economics drove the filtration technology improvements. We’re always looking to improve that optimisation between efficiency and service life and economics.”

With advances in filtration efficiency underway, he believes that the interest is to evolve the other two legs of the triangle, the pressure drop/service life economics that help move that equation along.

Benson also argues that the science behind still higher levels of filtration, covering particle below 0.1 microns is unclear. “There’s some debate in the industry over moving to a finer level of filtration. The literature is not clear on what the impact of that is on turbine performance. It may be they’re so small that they’re bouncing through, not attaching to the turbine blades.”

Retrofitting is widely considered to be a good opportunity for the industry. In the past, retrofitting tended to mean that pulse type systems would require modification to the filter house, but now there are new HEPA filtration products for retrofitting that do not require any filter houses modifications.

Emerging markets in Asia and the Middle East are also opening their eyes to high-efficiency media. For example, Nordic Air Filtration/TDC Filter’s high-efficiency TurboWeb® media has been chosen for a new project in China – the GT13E2 gas turbines at HPI’s Tongxiang plant in Zhejiang – due to the performance results shown globally.

Conversely, Garnett argues that the European market is slightly flat. But he acknowledges it does lead in terms of the usage of advanced filtration media. “The Europeans are much more advanced down this path now, roughly 5 per cent of the market has gone to HEPA filtration already. From 1 per cent or less 10 years ago,” says Garnett.

In any event, there is a clear market trend towards high-efficiency filtration and it will continue. As Garnett explains: “People want to optimise and the manufacturers of filters are reacting to that. You can’t stay still in this market place, change is constant.”

David Appleyard is a UK-based freelance journalist, specialising in energy.

More Power Engineering International Issue Articles
Power Engineering International Archives
View Power Generation Articles on PennEnergy.com